The overall objective of this phase of the research program is to modify in a predictable way specific genes in the germ line of experimental animals. i. Targetted modification in a planned way of the human beta globin locus was achieved in cultured cells during the last grant period. During the next, the first specific aim is to extend targetted gene modification to other selected loci. ii. The second specific aim is to achieve the same modifications in embryonic stem cells (ES cells) maintained in tissue culture on feeder layers. Correctly modified ES cells will be screened for their ability to differentiate when cultured without a feeder layer. iii. Modified ES cells that are still pluripotent will be injected into blastocysts for return to a foster mother and development into chimeric animals. Obtaining animals chimeric for the modified loci is the third specific aim. iv. The blastocyst injections will be repeated with different pluripotent clones of the modified ES cells until the fourth specific aim is reached - identifying chimeric animals that transmit the modifications through the germ line. Four target loci have been chosen either because of their amenability to targetting experiments or to provide needed animal models for significant human genetic diseases. I. The X- linked hypoxanthine phosphoribosyltransferase locus will be targetted first in somatic and then in ES cells using drug selection to identify cells in which the locus has been corrected or inactivated. II. The beta 2 microglobulin locus in mouse L cells will be modified to code for an immunologically distinct form of beta 2M on the cell surface. Factors affecting the efficiency of targetting this locus will be investigated. A null allele will be created in ES cells for germline transmission. III. In order to create a potential mouse model for sickle cell anemia, the beta D major and beta D minor genes at the mouse adult beta globin locus will be replaced by a single mouse/human hybrid gene carrying the human sickle cell mutation. IV. In order to create a small animal model of some aspects of human atherogenesis, the golden hamster apolipoprotein beta locus will be modified to mimic mutant pig apo B genes associated with coronary heart disease.